obsidian-tars
Obsidian tars plugin that supports text generation based on tag suggestions, using services like DeepSeek, Claude, OpenAI, OpenRouter, SiliconFlow, Gemini, Olla
What is obsidian-tars?
Obsidian tars plugin that supports text generation based on tag suggestions, using services like DeepSeek, Claude, OpenAI, OpenRouter, SiliconFlow, Gemini, Ollama, Kimi, Doubao, Qwen, Zhipu, QianFan & more.
Overview of obsidian-tars
What it is
Tars is an Obsidian plugin that supports text generation based on tag suggestions. It uses services like DeepSeek, Claude, OpenAI, OpenRouter, SiliconFlow, Gemini, Ollama, Kimi, Doubao, Qwen, Zhipu, QianFan and more. The name Tars comes from the robot Tars in the movie Interstellar. The plugin supports both desktop and mobile devices.
Key features
- Supports text generation triggered by tag suggestions (e.g., type
#+ tag + space). - Multimodal capabilities: image generation (GPT-Image-1) and visual understanding (image analysis via Claude, OpenRouter, SiliconFlow; PDF file analysis via Claude and OpenRouter).
- Tag commands: all tags are available in the command list; inserting tags based on selected sections or the section at cursor position.
- Custom prompt templates: run the Load template file command when using for the first time.
- Status bar displaying real-time information about character count, rounds, and time spent.
- Export conversations to JSONL dataset, supports ms-swift (Scalable lightWeight Infrastructure for Fine-Tuning).
- Support for internal links.
Use cases
- Generating AI responses within Obsidian notes by using tag-based commands.
- Creating structured conversations with AI assistants (system, user, assistant messages).
- Analyzing images and PDF documents embedded in notes.
- Generating and editing images using GPT-Image-1.
- Exporting conversation datasets for fine-tuning models.
Getting started
See the repository README.
License and pricing
The license is MIT. The plugin is open source.
Based on the project's public README and repository metadata.